# Northline Workshops — October 2026 results (CAD, tax excluded)

Period 2026-10-01 to 2026-10-31 inclusive. Rules: `inputs/policy.md`, "Northline Workshops accounting note — October 2026" (its only heading; paragraphs cited as para 1–3). Generated by `output/reconcile.py` from the three raw CSVs, which were read only.

## 1. Bottom line

- **9 paid orders; gross sales 2,820.00; applied refunds 300.00; net revenue 2,520.00; ad spend 450.00; contribution after advertising 2,070.00; net revenue per paid order 280.00.**
- **Course and combined net revenue, contribution and per-order figures are provisional.** Settled refund RF05 (row R06, order N1011, settled 2026-10-25) has no amount. Policy para 3 treats it as unknown, not zero, so it is left out of the calculable refund total. Workshop figures do not depend on it.
- Contribution after advertising is **not business profit**: no COGS, payment fees, salaries or overhead were supplied (policy para 1).

## 2. Results

| Measure | Workshop | Course | Combined |
|---|---:|---:|---:|
| Paid order count | 5 | 4 | 9 |
| Gross sales | 1,920.00 | 900.00 | 2,820.00 |
| Applied refunds (settled in October, calculable) | 240.00 | 60.00 ‡ | 300.00 ‡ |
| Net revenue = gross − applied refunds | 1,680.00 | 840.00 ‡ | 2,520.00 ‡ |
| Advertising spend | 300.00 | 150.00 | 450.00 |
| Contribution after advertising = net revenue − ad spend | 1,380.00 | 690.00 ‡ | 2,070.00 ‡ |
| Net revenue per paid order | 336.00 | 210.00 ‡ | 280.00 ‡ |

‡ Provisional: excludes the unknown amount of settled refund RF05 (R06).

- Workshop: 1,920.00 − 240.00 = 1,680.00; 1,680.00 − 300.00 = 1,380.00; 1,680.00 ÷ 5 = 336.00.
- Course: 900.00 − 60.00 = 840.00; 840.00 − 150.00 = 690.00; 840.00 ÷ 4 = 210.00.
- Combined: 2,820.00 − 300.00 = 2,520.00; 2,520.00 − 450.00 = 2,070.00; 2,520.00 ÷ 9 = 280.00.

**Limit on the provisional figures (a bound, not an estimate).** Policy para 3 says refunds cannot exceed paid gross, and N1011's paid gross is 360.00 (S13). If RF05 is valid, the final figures therefore lie between these values: Course net revenue 480.00 to 840.00; Course contribution 330.00 to 690.00; combined net revenue 2,160.00 to 2,520.00; combined contribution 1,710.00 to 2,070.00. The lower end is a full refund of N1011. The actual amount is not in the inputs and has not been guessed.

Included paid orders (`cleaned-sales.csv`):

- Workshop: N1001 (S01) 240.00; N1003 (S03) 480.00; N1005 (S06) 240.00; N1010 (S12) 720.00; N1012 (S14) 240.00
- Course: N1002 (S02) 120.00; N1006 (S07) 240.00; N1008 (S10) 180.00; N1011 (S13) 360.00

Applied refunds (`cleaned-refunds.csv`): RF01 (R01) 240.00 on N1003, Workshop, settled 2026-10-09; RF02 (R02) 60.00 on N1006, Course, settled 2026-10-14.

Included ad spend (`cleaned-ads.csv`): AD01 (A01) Workshop Search 150.00; AD02 (A02) Course Social 90.00; AD03 (A03) Workshop Social 100.00; AD04 (A05) Course Search 60.00; AD05 (A06) Workshop Search 50.00. By product and channel: Workshop Social 100.00; Workshop Search 200.00; Course Social 90.00; Course Search 60.00.

## 3. Reconciliation of included and excluded records

| Source file | Raw rows | Included (cleaned file) | Excluded (`exceptions.csv`) | Excluded row IDs |
|---|---:|---:|---:|---|
| inputs/sales.csv | 15 | 9 | 6 | S04, S05, S08, S09, S11, S15 |
| inputs/refunds.csv | 7 | 2 | 5 | R03, R04, R05, R06, R07 |
| inputs/ads.csv | 7 | 5 | 2 | A04, A07 |
| Total | 29 | 16 | 13 | |

Each raw row ID appears exactly once, in a cleaned file or in `exceptions.csv`. Cleaned files hold only records that enter the totals; flagged-but-unresolved records (R05, R06) are kept in `exceptions.csv` with their raw values.

| Amount tie-out (CAD) | All raw rows | Less excluded rows | = Included | Section 2 figure |
|---|---:|---:|---:|---|
| Sales gross | 4,140.00 | 1,320.00 | 2,820.00 | Gross sales 2,820.00 |
| Refund amounts | 960.00 + 1 blank | 660.00 + 1 blank | 300.00 | Applied refunds 300.00 |
| Ad spend | 630.00 | 180.00 | 450.00 | Ad spend 450.00 |

| Row | Record | Amount | Exception type | Effect on October totals |
|---|---|---:|---|---|
| S04 | N1003 | 480.00 | exact_duplicate | None - order counted once via retained row S03 |
| S05 | N1004 | 120.00 | status_pending | Not counted as a paid order or revenue |
| S08 | N1007 | 240.00 | status_cancelled | Not counted as a paid order or revenue |
| S09 | N1008 | 120.00 | superseded_conflicting_row | None - order counted once via S10 at gross 180.00 |
| S11 | N1009 | 240.00 | out_of_period | Not counted in October |
| S15 | N1013 | 120.00 | out_of_period | Not counted in October |
| R03 | RF02 on N1006 | 60.00 | exact_duplicate | None - refund counted once via retained row R02 |
| R04 | RF03 on N1005 | 240.00 | refund_pending | Not subtracted from October revenue |
| R05 | RF04 on N9999 | 120.00 | no_valid_paid_order | Not subtracted; no product can be assigned |
| R06 | RF05 on N1011 | (blank) | missing_refund_amount | Not subtracted (unknown, not zero); Course and combined net revenue are PROVISIONAL |
| R07 | RF06 on N1012 | 240.00 | settled_out_of_period | Not subtracted from October revenue |
| A04 | AD03 | 100.00 | exact_duplicate | None - ad counted once via retained row A03 |
| A07 | AD06 | 80.00 | out_of_period | Not counted in October ad spend |

`exceptions.csv` adds the explanation, policy reference, follow-up and the untouched raw row for each line above.

Policy checks that found nothing: no unmapped products; no applied refund exceeds its order's paid gross (RF01 240.00 against 480.00 on N1003; RF02 60.00 against 240.00 on N1006); the blank RF05 amount is the only empty or unparseable value.

## 4. Open items

1. **RF05 (R06), Course order N1011: amount missing.** Course and combined figures stay provisional until the settled amount is obtained (limits in section 2).
2. **RF04 (R05): settled 2026-10-19 for 120.00 against order N9999, which is not in `inputs/sales.csv`.** Flagged and excluded per policy para 2; no product can be assigned. If it proves to belong to a valid October paid order, October net revenue would fall by up to that amount; the inputs do not say which.
3. **RF03 (R04): pending 240.00 on Workshop order N1005.** Not subtracted (policy para 1). It will reduce net revenue in the month it settles.
4. **RF06 (R07): 240.00 on Workshop order N1012, settled 2026-11-02.** The order (dated 2026-10-31) counts in October; the refund belongs to November.
5. **N1008: two conflicting rows.** S10 (180.00, updated 2026-10-16T09:00:00) was retained over S09 (120.00) under the latest-updated_at rule.

## 5. Recommendation

- **Do not shift ad budget between products on these numbers.** Both products returned the same calculable net revenue per ad dollar: Workshop 1,680.00 ÷ 300.00 = 5.60; Course 840.00 ÷ 150.00 = 5.60. The Course figure is provisional and could be as low as 3.20 if RF05 is a full refund. The sample is 9 orders in one month, and the inputs do not link individual orders to ads.
- **Close the RF05 and RF04 questions before treating October as final.** They are the two flagged refund records that leave October's figures open.
- **Look into Workshop refunds first.** Three of the five paid Workshop orders have a refund record (N1003 applied, N1005 pending, N1012 settled in November). Applied plus pending plus November-settled Workshop refunds are 240.00 + 240.00 + 240.00 = 720.00, or 37.5% of October Workshop gross, against 12.5% recognised in October. Workshop's October net revenue of 1,680.00 does not reflect the 240.00 from RF06, which falls into November, or the 240.00 still pending on RF03. The inputs give no refund reasons, so the cause is not known.
- **Do not read 2,070.00 as profit.** It is net revenue less advertising only. COGS, payment fees, salaries and overhead were not supplied, so whether October was profitable cannot be determined from these files.

## 6. Method and assumptions

- Dates: ISO, DD/MM/YYYY slash form and English month names accepted, output as YYYY-MM-DD (policy para 3). `updated_at` timestamps kept as supplied.
- Money: `CAD` and `$` stripped, two decimals; blanks kept blank. Product and channel trimmed and case-normalised (Workshop/Course; Search/Social).
- Order of tests per ID: collapse exact duplicates (first row kept), then latest `updated_at` for conflicts, then status, period, product and amount.
- Assumption: a refund's product is that of its order in `cleaned-sales.csv`; `inputs/refunds.csv` has no product column.
- Assumption: a refund that is settled but dated outside October is classed as out-of-period rather than tested further (applies to R07).
- Assumption: 'valid paid order' means an order retained in `cleaned-sales.csv`. N9999 fails because it is absent from the sales file entirely.
- Net revenue per ad dollar in section 5 is an added ratio for the recommendation, not a policy-defined measure.

## 7. Checks performed by the script on this run

- Totals recomputed from the written cleaned CSVs, not from in-memory values; Workshop + Course = Combined for every additive measure.
- Row-ID reconciliation: every raw row appears once in a cleaned file or `exceptions.csv`; amount tie-outs in section 3 match section 2.
- All 21 figures in the section 2 table matched values worked separately by hand from the raw files.
